Genomics Goes to the Dogs

FIGURING FIDO

Beyond the content of the dog genome, these Hot Papers taken together represented one of the first examples of survey sequencing, in which a genome is sequenced with a relatively low level of redundancy and at a lower cost than more dense mapping projects. ?For people whose main interest in genome sequence projects is to be able to navigate back and forth between genomes, you didn?t really need to generate a 6x or 7x fully assembled sequence. With about 1.5x sequence and a dense radiation hybrid map, you could essentially get the same amount of information for a fraction of the time and a fraction of the cost,? says Ostrander.

Kirkness says that data from the survey sequencing have been used in a number of targeted studies on specific genes and diseases. Hannes Lohi of the Hospital for Sick Children in Toronto, and colleagues, for instance, used it in part to identify a gene for a specific form of epilepsy in miniature wirehaired dachsunds.4 ?What was really neat about that paper was that it described a kind of dodecamer repeat that was responsible for the disease ? and dodecamer repeats aren?t a disease mechanism we?ve ever seen in humans before,? says Ostrander, who has collaborated with Kirkness to analyze the extent of linkage disequilibrium in the dog genome and to compare haplotype sharing across breeds.5

The data have also been an asset to comparative genomic studies; for instance, Benson and colleagues used the papers as a resource to map a trait ? canine tricuspid valve malformation ? comparatively to syntenic regions in mouse, rat, and human.6 And, feline genome researcher William Murphy of Texas A&M University, and colleagues, used the radiation hybrid map data to help build a comparative map of several mammalian genomes to analyze evolutionary chromosomal breakpoint regions.7 ?Compared to other species, the [dog] map was one of the best quality maps at the time,? Murphy says.

THE BOXER IN BRIEF

Although the low-coverage map is widely cited as an excellent resource for these studies, dog genomics researchers and the NHGRI had recognized the need for a denser map. ?If you?re interested in the context of specific sequences ? where you?d like to know a complete gene structure for all of the exons, all of the introns, the controlling sequences, and you?d like that information for a large number of genes, then you would want a more complete picture,? says Kirkness.

The 7.5x boxer sequence3 was published in December 2005 and is publicly available through GenBank and EMBL-Bank, among other databases. Canine geneticist Patrick Venta of Michigan State University predicts that the differences between boxer and poodle sequence data will provide more SNP information. ?From the much deeper sequence, we?re able to learn a lot more about the amount of sharing of haplotypes that exist between different dog breeds. And that really facilitates designing studies to do whole-genome association studies,? adds Ostrander (a coauthor on the boxer paper).

The initial work, although not mainstream for genomics, continues to thrive, and one possible reason has been the strong commitment from the dog-owner community. For those interested in keeping tabs on good breeding, a genomic blueprint could be invaluable. Ostrander and Galibert?s research was funded in part by the American Kennel Club, for example. Ostrander says that dog lovers both in and out of the scientific community are passionate about their animals (she herself owns a border collie named Tess) and are willing to help in any way they can. ?We have freezers full of thousands of cheek swabs provided by dog owners from our attendance at dog shows. That?s been a great way to do science.

References